Page 1 of 2 12 LastLast
Results 1 to 15 of 18

Thread: Just found: "Blurred Out: 51 Things You Aren't Allowed to See on Google Maps"

  1. #1
    Member Fritriac's Avatar
    Join Date
    Oct 2007
    Location
    50░35'2.06"N, 8░31'51.24"E
    Posts
    37

    Default Just found: "Blurred Out: 51 Things You Aren't Allowed to See on Google Maps"

    I just found this article over Fark.com

    Just if you're intereested in it ...
    ... but never eat the yellow snow! (Inuit wisdom)

  2. #2
    Junior Member
    Join Date
    Oct 2008
    Posts
    1

    Default i found somthing else

    Goto this location...


    .... Latitude 57░52'6.00"N
    Longitude 8░28'59.19"W

    it shows an island that is half faded.

    ha ha

  3. #3
    Junior Member
    Join Date
    Oct 2008
    Posts
    6

    Default

    What's going on with the lines of longitude there? They are all kinked on my GoogleEarth 4.3.7284.3916 (beta).

    Take a look at 55░N 137░E, Feklistova Island. Why is the island covered in a green blur? Green blurring seems popular in this region of Khabarovsk Krai. If you zoom out a little and look at the coastline of the mainland due South of Feklistova, there is a extensive green blurring of what is ostensibly the sea, yet the coastline overlay locates the shoreline to the East.

  4. #4
    Super Moderator Munden's Avatar
    Join Date
    Aug 2006
    Location
    Maryland, USA
    Posts
    939

    Default

    Quote Originally Posted by chemist View Post
    What's going on with the lines of longitude there? They are all kinked on my GoogleEarth 4.3.7284.3916 (beta).

    Take a look at 55░N 137░E, Feklistova Island. Why is the island covered in a green blur? Green blurring seems popular in this region of Khabarovsk Krai. If you zoom out a little and look at the coastline of the mainland due South of Feklistova, there is a extensive green blurring of what is ostensibly the sea, yet the coastline overlay locates the shoreline to the East.
    Nothing is 'blurred' other than it's just even lower resolution imagery than the Landsat that was used to create most of the globe they now use as the base image. You can consider that island in the lowest resolution default globe image. Notice that white halo around the dark peninsula of the island next to it? If you zoom in you can see that the halo is actually just the water in the higher resolution Landsat image which fades rapidly into the blobby blue of the other image.

    It's difficult to put properly in words, but in summary it's not blurred, it was just never raised to the higher resolution Landsat globe image.

  5. #5
    Junior Member
    Join Date
    Oct 2008
    Posts
    6

    Default

    Quote Originally Posted by Munden View Post
    Nothing is 'blurred' other than it's just even lower resolution imagery than the Landsat that was used to create most of the globe they now use as the base image. You can consider that island in the lowest resolution default globe image. Notice that white halo around the dark peninsula of the island next to it? If you zoom in you can see that the halo is actually just the water in the higher resolution Landsat image which fades rapidly into the blobby blue of the other image.
    "fade"? if you paste two images together there the boundary would be a line, would it not? You'd use the high res pixels until they ran out. Actually there is a sharp, roughly latitudinal, line cutting the West coast of Bolshoy Shantar Island at about 54░58'N, but the transition to green blur spans this line.

    You don't address the peculiar coastline to the south. It looks to me as though someone has tried to combine a low res and a high res image of this area and got the registration wrong.

  6. #6
    Senior Member tpstigers's Avatar
    Join Date
    Feb 2008
    Location
    New England
    Posts
    111

    Default

    This '51 things blah blah' made the rounds months ago. Shortly thereafter, the truth of the matter made the rounds. If you're interested, a decent article on the subject can be found here.
    I have an existential map; it has 'you are here' written all over it.

  7. #7
    Super Moderator Munden's Avatar
    Join Date
    Aug 2006
    Location
    Maryland, USA
    Posts
    939

    Default

    Quote Originally Posted by chemist View Post
    "fade"? if you paste two images together there the boundary would be a line, would it not? You'd use the high res pixels until they ran out. Actually there is a sharp, roughly latitudinal, line cutting the West coast of Bolshoy Shantar Island at about 54░58'N, but the transition to green blur spans this line.

    You don't address the peculiar coastline to the south. It looks to me as though someone has tried to combine a low res and a high res image of this area and got the registration wrong.
    Fade, yes. In addition to the bit of imagery on the island you point out, if you look around you will see hundred of instances where images faded between two different versions of the image. Keep in mind we see all of these 'ghost ships' in harbors where there was an image seam. There would be no ghost ships if they simply did a hard image boundary stitch. Why they do the fade, I don't know for sure, but they do it frequently. I suspect it makes it easier for them to stitch images together, but that's just my guess.

    The docks along the western side of Manhattan still show the ghosting from fading one image to the other so that there is clearly a transparency showing both images on the boundary.

    As for the southern area looks like a matter of someone doing up a coastline border based on the low resolution imagery. What exactly got messed up is anyone's guess, but it looks to me like the higher resolution image is correct and the blurry green lower resolution is wrong, and the border outline is based on the lower resolution. Either resolution could technically be wrong, correct. I just can't find strange stitching errors in the high resolution imagery, but it could still be warped when it was wrapped onto the globe. We'll have to see where they put imagery tiles when they get around to updating the area to proper high resolution imagery.

  8. #8
    Super Moderator Munden's Avatar
    Join Date
    Aug 2006
    Location
    Maryland, USA
    Posts
    939

    Default

    Quote Originally Posted by tpstigers View Post
    This '51 things blah blah' made the rounds months ago. Shortly thereafter, the truth of the matter made the rounds. If you're interested, a decent article on the subject can be found here.
    heh, well the original post in this thread was back almost two months ago.

  9. #9
    Junior Member
    Join Date
    Oct 2008
    Posts
    6

    Default

    Quote Originally Posted by Munden View Post
    Fade, yes. In addition to the bit of imagery on the island you point out, if you look around you will see hundred of instances where images faded between two different versions of the image. Keep in mind we see all of these 'ghost ships' in harbors where there was an image seam. There would be no ghost ships if they simply did a hard image boundary stitch. Why they do the fade, I don't know for sure, but they do it frequently. I suspect it makes it easier for them to stitch images together, but that's just my guess.

    The docks along the western side of Manhattan still show the ghosting from fading one image to the other so that there is clearly a transparency showing both images on the boundary.

    As for the southern area looks like a matter of someone doing up a coastline border based on the low resolution imagery. What exactly got messed up is anyone's guess, but it looks to me like the higher resolution image is correct and the blurry green lower resolution is wrong, and the border outline is based on the lower resolution. Either resolution could technically be wrong, correct. I just can't find strange stitching errors in the high resolution imagery, but it could still be warped when it was wrapped onto the globe. We'll have to see where they put imagery tiles when they get around to updating the area to proper high resolution imagery.
    Other imagery, eg Microsoft VEarth, seem to put the coastline where the low res image (green blur) ends, suggesting that it's the high res image that google got wrong. A whole small river system that discharges just to the west of that peninsular (which is bright white in GE) is present on VE but missing from GE. Immediately to the North of Medvezhiy Is. on GE there is a "ghost" coastline, and there are ghost rivers running through the sea ice and even the sea to the east of that island.

    I hope the military don't use Google Earth for targeting.

  10. #10
    Junior Member
    Join Date
    Oct 2008
    Posts
    6

    Default

    53░15'N 141░40'E ghost land in the sea or clouds?

  11. #11
    Super Moderator Munden's Avatar
    Join Date
    Aug 2006
    Location
    Maryland, USA
    Posts
    939

    Default

    Quote Originally Posted by chemist View Post
    53░15'N 141░40'E ghost land in the sea or clouds?
    Neither. It's shallow water. The swirling patterns in the water are silt and 'sand' bars. There's also a lot of sea foam in areas, which is the what the white lines are. The regularly spaced white lines under the placemark look like foam, but might be related to large waves forming over a particularly shallow bit of water.

  12. #12
    Super Moderator Munden's Avatar
    Join Date
    Aug 2006
    Location
    Maryland, USA
    Posts
    939

    Default

    Quote Originally Posted by chemist View Post
    Other imagery, eg Microsoft VEarth, seem to put the coastline where the low res image (green blur) ends, suggesting that it's the high res image that google got wrong. A whole small river system that discharges just to the west of that peninsular (which is bright white in GE) is present on VE but missing from GE. Immediately to the North of Medvezhiy Is. on GE there is a "ghost" coastline, and there are ghost rivers running through the sea ice and even the sea to the east of that island.

    I hope the military don't use Google Earth for targeting.

    Heh, there are indeed some screwups in some of the data. The very low resolution isn't really necessarily a Google issue. I've seen other areas where a problem in the low res image was actually in more than one program. Virtual Earth, for instance, shared a glitch in the data with Google until GE updated to a satellite image there.

    37░51'18.07"S, 175░39'4.39"E

    If you visit that in VE you'll see that there are two strips of image data that are duplicated. They're most easily visible on the souther edge of the woods.
    They had the same original low res source though.

    I'm looking harder at the coastal area you pointed out and I actually do see the problem now. Whoever slapped that data together has indeed shifted the coast to the west, you're absolutely correct. It's transparent so you can see the proper terrain through it, though, and it just fades out so which is why the rest of the terrain ends up being correct as you move to the west.

    I realized that sharp peninsula isn't white, but it is showing through the gleaming white water of the coastal imagery. Then going slightly west to the label 'Ostrov Medvezhiy' you can very clearly see the two coastlines are just laid over on top of each other. The dark green coastal land has a pale white coast on top of it. (ice?)

  13. #13
    Junior Member
    Join Date
    Oct 2008
    Posts
    6

    Default

    Quote Originally Posted by Munden View Post
    Neither. It's shallow water. The swirling patterns in the water are silt and 'sand' bars. There's also a lot of sea foam in areas, which is the what the white lines are. The regularly spaced white lines under the placemark look like foam, but might be related to large waves forming over a particularly shallow bit of water.
    Yes, I think you are right. I wasn't expecting to see structure in the sea, but looking closer it's definitely not land.

    Quote Originally Posted by Munden
    Heh, there are indeed some screwups in some of the data. The very low resolution isn't really necessarily a Google issue. I've seen other areas where a problem in the low res image was actually in more than one program. Virtual Earth, for instance, shared a glitch in the data with Google until GE updated to a satellite image there.

    37░51'18.07"S, 175░39'4.39"E
    I was using the web interface at live.com.au which doesn't seem to understand lat/long.


    Quote Originally Posted by Munden
    If you visit that in VE you'll see that there are two strips of image data that are duplicated. They're most easily visible on the souther edge of the woods.
    They had the same original low res source though.

    I'm looking harder at the coastal area you pointed out and I actually do see the problem now. Whoever slapped that data together has indeed shifted the coast to the west, you're absolutely correct. It's transparent so you can see the proper terrain through it, though, and it just fades out so which is why the rest of the terrain ends up being correct as you move to the west.

    I realized that sharp peninsula isn't white, but it is showing through the gleaming white water of the coastal imagery. Then going slightly west to the label 'Ostrov Medvezhiy' you can very clearly see the two coastlines are just laid over on top of each other. The dark green coastal land has a pale white coast on top of it. (ice?)
    It can be difficult to tell the difference between cloud, snow, and sea ice, plus this area looks very different between summer and winter. This may have been the cause of the confusion. Are the source images reconciled manually or is it all done automatically?

  14. #14
    Super Moderator Munden's Avatar
    Join Date
    Aug 2006
    Location
    Maryland, USA
    Posts
    939

    Default

    Quote Originally Posted by chemist View Post
    I was using the web interface at live.com.au which doesn't seem to understand lat/long.
    whoops, sorry about the lat/long thing. Reading the maps.live.com URL implies these are the coordinates: -37.849171, 175.649071

    The duplicated bits of the image are easily seen at that location and then to the west where there's another batch of forest.

    Quote Originally Posted by chemist View Post
    It can be difficult to tell the difference between cloud, snow, and sea ice, plus this area looks very different between summer and winter. This may have been the cause of the confusion. Are the source images reconciled manually or is it all done automatically?
    When it comes to the low resolution imagery I really have no idea unfortunately. I would imagine it's a little like other imagery in that the satellite might store some meta-data within the image about the coordinates it thinks match the boundaries of the image. I'm sure those are prone to some error though, and most of the imagery is manually manipulated to fit correctly. That's been my assumption.

  15. #15
    Junior Member
    Join Date
    Oct 2008
    Posts
    6

    Default

    Quote Originally Posted by Munden View Post
    whoops, sorry about the lat/long thing. Reading the maps.live.com URL implies these are the coordinates: -37.849171, 175.649071

    The duplicated bits of the image are easily seen at that location and then to the west where there's another batch of forest.
    It looks like a strip has been replaced with a duplicate of the imagery immediately to the South, almost from coast to coast!

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •